An example of a two-sided linear pulse motor conventionally used in an FA device or exposure apparatus includes a 4-phase linear pulse motor with an arrangement as shown in FIG. 4. Referring to FIG. 4, first and second primary stators 1 are arranged to sandwich a secondary movable element 2. Excitation units 3a, 3b, 3c, and 3d respectively classified in A, B, C, and D phases and made of a magnetic material are arranged along the traveling direction of the secondary movable element 2. The excitation units 3a, 3b, 3c, and 3d respectively have excitation coils 4a, 4b, 4c, and 4d wound around them, and projecting primary stator pole teeth 5a, 5b, 5c, and 5d each formed at least one on their respective surfaces opposing the secondary movable element 2.
The secondary movable element 2 has a secondary movable element yoke 6 extending continuously in the traveling direction of the secondary movable element 2, and a plurality of secondary movable element pole teeth 2a arranged on the two side surfaces of the secondary movable element yoke 6 to oppose the primary stators 1. The secondary movable element 2 is supported to be movable in the traveling direction of the secondary movable element 2 with respect to the primary stators 1. To drive the secondary movable element 2, power is supplied to the 4-phase excitation coils 4a, 4b, 4c, and 4d in a certain order so the excitation units 3a, 3b, 3c, and 3d of the respective phases are excited. The secondary movable element 2 obtains a thrust in the traveling direction from a magnetic attracting force generated by the mutual magnetic functions of the movable element pole teeth 2a and stator pole teeth 5a to 5d. 
In the conventional two-sided linear pulse motor described above, the secondary movable element 2 has the plurality of secondary movable element pole teeth 2a on those two side surfaces which oppose the first and second primary stators 1 from the secondary movable element yoke 6 extending in the traveling direction. Since the secondary movable element yoke 6 becomes large in size, the weight of the entire secondary movable element 2 increases, and the inertia of the movable element increases. Consequently, even when a control command for sharp acceleration/deceleration is supplied to the linear pulse motor, the linear pulse motor cannot reach the specified speed within a predetermined period of time, or cannot be decelerated to stop within a predetermined period of time. Namely, this linear pulse motor has a poor response performance, i.e., a poor accelerating/decelerating performance.
As a conventional VR linear pulse motor, for example, a 4-phase linear pulse motor as shown in FIG. 10 is available. In this linear pulse motor, a movable element 11 has magnetic portions 13a, 13b, 13c, and 13d respectively classified in A, B, C, and D phases and formed along the traveling direction of the movable element 11, coils 14a, 14b, 14c, and 14d wound around the corresponding magnetic portions, and projecting movable element pole teeth 15a, 15b, 15c, and 15d each formed at least one on that surface of the movable element 11 which opposes a stator 60. The movable element 11 is supported to be movable with respect to the stator 60. The stator 60 has a plurality of stator pole teeth 12a on one side surface of a stator yoke 20 which opposes the movable element pole teeth 15a to 15d. To drive the movable element 11, power is supplied to the 4-phase excitation coils 14a to 14d in a certain order so the excitation units 13a, 13b, 13c, and 13d of the respective phases are excited. The movable element 11 obtains a thrust in the traveling direction from a magnetic attracting force generated by the mutual magnetic functions of the movable element pole teeth 15a to 15d and stator pole teeth 12a. In this case, assume that when power is supplied to the coil 14a, 14b, 14c, or 14d, this is called A-, B-, C-, or D-phase excitation.
The conventional excitation method includes three types of excitation, i.e., 1-phase excitation, 2-phase excitation, and 1-2-phase excitation. In the following description, the thrust generating principle will be explained by using 1-phase excitation as an example, and its problems will be discussed. In 1-phase excitation, the excitation phase is switched in the order of A phase→B phase→C phase→D phase→A phase, and during this excitation switching, the movable element can travel for 1 pitch P in the traveling direction. FIGS. 11A to 11C show how the movable element travels for ¼ pitch. P (pitch) mentioned here refers to the distance from a stator pole tooth to an adjacent stator pole tooth.
First, the movable element is energized in the A phase, so it forms a main magnetic circuit (broken line) as shown in FIG. 11A and maintains a stable state at this position. A stable state means a state wherein a thrust generated by the magnetic attracting force caused between the movable element pole teeth 15b and 15d and the stator pole teeth 12a upon A-phase excitation is balanced in the traveling direction. When excitation phase is switched from the A phase to the B phase, concerning the main magnetic circuit to be formed, in addition to the one which forms in the order of the magnetic portion 13b→magnetic portion 13a→movable element pole tooth 15a→stator pole tooth 12a→stator yoke 20→stator pole tooth 12a→movable element pole tooth 15b→magnetic portion 13b, the one which forms in the order of the magnetic portion 13b→magnetic portion 13d→movable element pole tooth 15d→stator pole tooth 12a→stator yoke 20→stator pole tooth 12a→movable element pole tooth 15b→magnetic portion 13b (FIG. 11B) is obtained. Of the magnetic fluxes that pass through the magnetic portions 13a, 13b, 13c, and 13d, the maximum one is that of the magnetic portion 13b. Regarding the thrust generated when the movable element pole teeth receives a magnetic attracting force from the stator pole teeth, in relation to the positions of the movable element pole teeth and stator pole teeth relative to each other, the thrust is zero in the movable element pole tooth 15a, a large leftward thrust is generated in the movable element pole tooth 15b, and a small rightward thrust is generated in the movable element pole tooth 15d. Accordingly, the movable element becomes unstable. Upon obtaining the rightward thrust, the movable element 11 moves to a position where the leftward and rightward thrusts are balanced, as shown in FIG. 11C. After this, the C, D, and A phases are excited sequentially.
A thrust is generated in the linear pulse motor, while the opposing movable element pole teeth and stator pole teeth are displaced from each other in the traveling direction, when a magnetic flux flows from one pole tooth to an opposing pole tooth through the gap and the stator pole teeth supplies a magnetic attracting force to the movable element pole teeth. In the conventional linear pulse motor, concerning the magnetic circuit shown in FIG. 11B, the magnetic flux generated by B-phase excitation passes between the movable element and stator through the gap at the portions of the movable element pole teeth 15a, 15b, 15c, and 15d. However, the magnetic flux that actually contributes to generation of the thrust generates a thrust only in the movable element pole tooth 15b. In the movable element pole teeth 15a and 15d, the magnetic flux does not generate a thrust, or generates a thrust in a direction opposite to the traveling direction. When a gap is present in the magnetic circuit, a magnetic resistance is present accordingly, and a magnetomotive force by excitation is necessary. When a gap that does not generate a thrust or a gap that generates a thrust in a direction opposite to the traveling direction is present in the magnetic circuit, a magnetomotive force is necessary accordingly, and the conversion efficiency from the magnetomotive force into the thrust becomes poor.